In the wake of the explosive scandal surrounding Michael LaCour, the UCLA political science grad student who faces overwhelming evidence that he fabricated the data behind a highly celebrated, widely publicized study about how to shift gay-marriage attitudes published in Science last December, the scientific community has a rare opportunity to reconsider its obsession with “clever young things” — and the professional pressures that create (and corrupt) them.
It’s tempting to argue, as some social scientists have, that the LaCour scandal is proof that the system works. After all, LaCour was caught, and it’s likely he will be drummed out of academia entirely. The Science paper was quickly retracted once its author’s fraud came to light. But if academics are satisfied to view this incident as an anomaly that was effectively handled by a functional system of checks, it would be a regrettable missed opportunity. Because while scandals like this one are (mercifully) rare, the underlying issues in academia that likely fueled LaCour’s misdeeds and delayed their discovery need to be addressed.
It’s worth pointing out that this isn’t just an academic problem — these issues ripple into the real world. At a time when social science research findings enjoy increasing coverage in big-time news outlets as well as dedicated sections in scientific magazines and niche blogs alike, and when social scientists themselves have the ear of government officials like never before, this research is indirectly — and sometimes directly — shaping all manner of public policy, as well as greatly influencing the vocabulary laypeople use to explain why humans do what they do. In other words, maybe you don’t think you’re interested in obscure, distant-seeming academic scandals and debates. But if you care about the gender pay gap, the police’s relationship with young black men, the financialization of the economy, or any of dozens of other issues that social scientists take up, then you have a stake in how social science is conducted.
With that in mind, there are at least three profitable conversations about academia that the LaCour scandal should provoke:
First, we need to figure out how to realistically slow down the ultracompetitive world of academia for young researchers. It’s been clear for a while now that there are far more Ph.D.s produced each year than there are job openings (especially in the social sciences and humanities), and to get one of those jobs a graduate student needs to publish a lot and in high-end journals. The demand has become so excessive that one professor recently proposed that journals stop accepting manuscripts from grad students to stave competition and level the playing field (not likely).
The profound pressure to publish certainly can’t explain LaCour’s deception on its own — every day, countless graduate students slog through imperfect data and tepid findings without engaging in fabrication. Still, the ever-intensifying contest for jobs and for recognition that is inherent to being a young social-science researcher makes these sorts of catastrophes — as well as all sorts of other, more subtle forms of misconduct — far more likely. And while the academic community has begun to take these pressures seriously in recent years, practical dialogues about how to find a better way are rare.
Second, we need some much-needed candor regarding just how easy it is to falsify social-science data. Even as young researchers face demands to publish splashy results in big journals, most conduct their research with notably little oversight. While LaCour’s alleged fabrication is a particularly egregious one, all researchers face the temptation to add or delete a few lines of data in their spreadsheets so as to firm up their statistical argument, or to “paraphrase” an interviewee’s quotation in a way that makes them say what the researcher needs them to say.
Academics need to confront the fact that “minor” infractions of integrity lead to “findings” that have the potential to change public policy, redirect resources, and transform people’s behavioral choices in potentially harmful ways. When our publications are based on incomplete representations of data, it brings our work closer to the style of cherry-picked, anecdotal arguments deployed in partisan media to which academic scholarship is supposed to be an antidote.
The current attempts to instill in young researchers a spirit of academic integrity are often half-baked. Some graduate programs include a perfunctory unit on “research ethics” somewhere in their early curricula, but that conversation tends to happen at a moment in students’ graduate careers before they’ve started to work with their own data and had a chance to feel the temptation to tweak their data when it isn’t telling the story they want it to. As a result, the conversation about data falsification in graduate school can be abstract and unmemorable — if it happens at all.
Third, we need to rethink the circumstances under which it is appropriate for senior faculty to be named as co-authors on studies that they did not conduct. One of the reasons LaCour almost got away with an extravagant lie is that his ostensible co-author, the highly regarded Columbia political scientist Donald Green, was across the country and played no role in administering the — apparently, in retrospect, nonexistent — surveys that were the cornerstone of his and LaCour’s results. (Green, who is known within political science for his focus on experimental design and his tendency to take on many mentees, quickly asked the paper be retracted after the fraud was revealed, and then expressed embarrassment and regret about the whole thing.)
It may seem strange to non-academics, but the cultural practice of giving senior faculty members co-authorship of papers to which they contributed supervision rather than substantive intellectual work has been commonplace since at least World War II. It’s a self-perpetuating system that professionally benefits both the junior scholar, who can claim a collaborative relationship with a well-regarded faculty member, and the senior scholar, who can give off the appearance of greater research productivity.
This habit isn’t likely to change anytime soon, so it’s important to clarify and underscore the responsibilities of senior collaborators. Faculty members are encouraged to treat their graduate student collaborators and advisees like colleagues rather than underlings, and mutual trust is a major part of that relationship. To move toward a culture in which senior faculty fact-checked every aspect of their students’ research would be to debase that trust. So the solution shouldn’t be to treat every graduate student like they’re the next Michael LaCour, but rather to inject these relationships with a little bit of skepticism — an ideal that is, after all, supposed to be one of the driving forces of academia. There need to be frank discussions in faculty meetings across the country about how easy it is to make stuff up and how to reconcile that fact with a necessary faith in one another’s professional integrity.
As the immediate aftermath of the LaCour scandal fades a bit, academics face a choice: They can see this as an unfortunate, one-off affair and get back to their own work, or they can take this opportunity to confront the systemic forces that lead to bad social science. It’s clear which of these options would entail more hard work and more painful conversations. It’s equally clear which one is right.
Drew Foster is a doctoral candidate in sociology at the University of Michigan.